Sciencedirect.Com

Home , Annexin A5, Osteonectin

d e

n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193

Available online at www.sciencedirect.com

ScienceDirect

jo urnal homepage: www.intl.elsevierhealth.com/journals/dema

Effect of the degree of conversion of resin-based

composites on cytotoxicity, cell attachment, and

gene expression

a,1 b,1 c

Masako Fujioka-Kobayashi , Richard J. Miron , Adrian Lussi ,

d e f,1

Reinhard Gruber , Nicoleta Ilie , Richard Bengt Price , b,g,∗,1 Gottfried Schmalz

Department of Cranio–Maxillofacial Surgery, Inselspital, Bern University Hospital, University of Bern,

Freiburgstrasse, Bern, 3010, Switzerland

Department of Periodontology, School of Dental Medicine, University of Bern, Freiburgstrasse 7, Bern, 3010, Switzerland

Department of Preventive, Restorative and Pediatric Dentistry, School of Dental Medicine, University of Bern,

Freiburgstrasse 7, Bern, 3010, Switzerland

Department of Oral Biology, Dental School, Medical University of Vienna, Sensengasse 2a, Vienna, 1090, Austria

Department of Conservative Dentistry and Periodontology, University Hospital, LMU Munich, Goethestr 70,

Munich, 80336, Germany

Department of Dental Clinical Sciences, Faculty of Dentistry, Dalhousie University, 5981 University Ave., Halifax,

Nova Scotia, B3H 4R2, Canada

Department of Conservative Dentistry and Periodontology, University Hospital Regensburg, University of

Regensburg, Regensburg, 93042, Germany

a r t i c l e i n f o a b s t r a c t

Article history: Objective. This study investigated the inﬂuence of the degree of conversion (DC), resin-based

Received 12 April 2019 composites (RBC) composition, and the effect of additional violet light from one light curing

Accepted 14 May 2019 unit (LCU) on cell attachment/growth, eluate cytotoxicity, and gene expression.

Methods. The effect of different DC of RBCs on human gingival ﬁbroblasts (HGFs) when cul-

tured directly onto cured RBCs, and when exposed afterwards to eluates in cell culture

® ®

Keywords: medium was examined. Venus (RBC-V; Bis-GMA-based) and Venus Pearl (RBC-P; TCD-DI-

Resin-based composites HEA and UDMA-based) were cured using a single emission peak (blue) light, Translux Wave ;

Light curing unit TW and a dual emission peak (blue-violet) light, Translux 2 Wave ; T2W. To determine the

Resin conversion value of the additional violet light from the T2W, exposure times and distances were adjusted

Cytotoxicity to deliver similar radiant exposures (RE) from the blue region of both lights at ﬁve different

2 2

RE levels from 1.5 J/cm to 28.9 J/cm .

∗

Corresponding author at: Department for Conservative Dentistry and Periodontology, University Hospital Regensburg, Regensburg, D-

93042, Germany.

E-mail address: [email protected] (G. Schmalz).

https://doi.org/10.1016/j.dental.2019.05.015

1174 d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193

Results. Both RBCs light-cured with the T2W at higher REs resulted in higher DC, increased

cell adhesion and decreased eluate cytotoxicity. RBC-V induced greater cell adhesion, lower

mRNA levels of pro-inﬂammatory markers, and higher mRNA levels of a proliferation marker

than RBC-P. Wettability was the same for both RBCs. Toxicity decreased with increasing

number of elution cycles. The initial eluates from RBC-P had a lower toxicity than from

RBC-V.

Signiﬁcance. RBCs cured with T2W (delivering both blue and violet light) at higher RE had

greater DCs. The greatest DC and the least cell reactions were observed when the RE was 2

>25 J/cm .

increased solubility and water sorption due to the reduced

1. Introduction

conversion of the monomers [14]. One factor that inﬂuences

the DC of light cured RBCs is the radiant exposure (RE; dose

Resin-based composite (RBC) restorations are routinely used

or amount of energy delivered per unit area) and the wave-

in dental practice [1,2], and their use will continue to rise with

lengths of light that are delivered to the RBC. Accordingly,

the global phase down in the use of amalgam as part of the

the cytotoxicity of RBCs and their respective dental adhesives

Minamata Convention on Mercury [3]. The clinical success

has been reported to be dependent upon the LCU; [15,16] a

of these restorations requires that the RBCs achieve sufﬁ-

high power quartz-tungsten halogen LCU that was claimed

cient mechanical properties and exhibit good biocompatibility 2

to deliver an irradiance of 3000 mW/cm produced a signiﬁ-

in the oral cavity. Despite their widespread use, delivering a

cantly higher cell viability of some RBCs compared to when

successful RBC in deep proximal cavities continues to be a

a quartz-tungsten-halogen LCU that was claimed to deliver

challenge, partly because the RBC at the bottom of the approx- 2

an irradiance of 600 mW/cm was used. However, in other

imal cavity ﬂoor is often 7 mm or more away from the tip

studies, no difference in the DC and cytotoxicity of dental

of the light curing unit (LCU), and this region is often difﬁ-

adhesives were found when the distance between light tip

cult to access with the curing light. Also, as the thickness

and material was increased up to 7 mm [17]. No systematic

increases, there is a logarithmic decrease in the light trans-

studies on the effect of DC and emission spectrum from the

mission through both the RBC and the tooth, further reducing

LCU on cytotoxicity of RBCs are available. Such studies are

the amount of light that reaches the bottom of the restoration

necessary because the DC depends more on the RE, and the

[4]. Thus, the RBC at the bottom of the proximal box is often

emission spectrum from the LCU and less upon the irradi-

less well polymerized than the top surface of the RBC.

ance value alone. Since a high irradiance delivered at the

RBCs are complex mixtures of inorganic ﬁller particles,

wrong wavelength will not cure the resin, an appropriate

matrix resins, coupling agents and additives. The extent to

combination of wavelengths from the LCUs that match the

which the monomers are converted to a polymer, the degree

absorption characteristics of the photoinitiators is required

of conversion (DC) of the functional groups of the RBC, has

to achieve a satisfactory DC. This was not an issue when

been reported to range from 35 to 77% [5]. In the mouth,

using quartz-tungsten halogen LCUs that emit a broad spec-

the RBCs will release some of their unreacted monomer

trum of ﬁltered light from 380 to 515 nm. However, due to

content and other substances [5,6] and it is estimated that

the nature of the light emitting diode (LED) emitter, LED-

approximately 2 wt.% of the organic matrix; e.g., Bis-GMA

curing lights do not produce such a broad emission spectrum

(0.4 wt.%–1.5 wt.%), TEGDMA (0.04–2.3 wt.%), is elutable from

unless they include multiple different LED emitters to pro-

the tested RBC in aqueous media [7–9]. The amount of elutable

vide multiple emission peaks to cover between 380–515 nm.

substances affects the biocompatibility of the RBCs for it is

This is unlike the single emission peak LCUs that emit just

known that some monomers or compounds eluted from com-

a single emission peak that is usually between 450 and

posites may cause both local and systemic adverse reactions

475 nm [6]. The manufacturers of these multiple peak wave-

[10]. The greatest release of these elutable substances (e.g.,

length LCUs claim that these broad-spectrum LEDs cure all

HEMA, TEGDMA, Bis-GMA, UDMA) takes place within the ﬁrst

RBCs containing a variety of photoinitiators, such as 2,4,6-

few hours after photo-curing, and it declines asymptotically

trimethylbenzoyldiphenylphosphine oxide (Lucirin TPO) or

over time if the incubation solution is not refreshed [10,11].

phenylpropanedione (PPD), whose absorption peak is close

However, even after 90 and 180 days, refreshed solutions still

to 390 nm, as well as camphorquinone (CQ), whose peak

contain elution products (e.g., TEGDMA, Bis-GMA, Bis-EMA,

absorbance is close to 470 nm [6,18].

BPA) from several brands of RBC [12].

When the RBC is placed into sub-gingival regions, it

The DC of RBCs affects the release of potentially toxic

interacts with epithelial cells and ﬁbroblasts from gingival

substances that are present within the RBC [13]. Inadequate

tissues. In general, a good attachment of these cells to bio-

polymerization of the RBCs has been shown to result in

materials is desired to support cell spreading, proliferation

and new tissue formation [19,20]. The adhesive behavior of

osteoblasts or ﬁbroblasts to model surfaces is dependent upon

These authors contributed equally to this work.

d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193 1175

the chemistry of these surfaces [20,21]. Schweikl et al. used

Table 1 – Material speciﬁcations (information supplied

self-assembled monolayers that were terminated by various

by Kulzer GmbH).

functional chemical groups to demonstrate that cell prolifera-

Composite, Lot # Composition

tion on hydrophobic surfaces, such as n-octyltriethoxysilane,

shade

can be as high as the cell proliferation that occurs on moder-

Venus A3 010036A BIS-GMA matrix, 58,7 % ﬁller by

ately hydrophobic surfaces and hydrophilic oxidized surfaces

(RBC-V) volume, which is Barium Aluminium

[20]. Moreover, the cells attached to biomaterials showed dif-

Fluoride glass (∅ 0.7 ␮m; max. < 2 ␮m) •

ferent gene expression patterns. It was reported that human

Highly dispersive Silicon Dioxide (∅

osteoblasts cultured on titanium or zirconia expressed mRNAs

0.04 ␮m); photoinitiator system:

differently for speciﬁc markers including Runx-2, -3, BMP-7, camphorquinone (CQ) / amine

alkaline phosphatase, osteopontin, osteocalcin, osteonectin, Venus Pearl 010504A TCD-DI-HEA and UDMA,

A3 approximately 59% ﬁller by volume,

Type I collagen, bone sialoprotein and integrin ␤3, primarily

(RBC-P) with 58% organic ﬁller by volume and

due to small differences in the roughness and marginally due

a particle size of 5 nm – 5 ␮m • Barium

to the material composition in the initial phase of attachment

Aluminium Fluoride glass •

and proliferation [22,23]. Although several studies have inves-

Pre-polymerized ﬁller • Highly

tigated the effect of the RBC on cell behavior by testing eluates

discrete nanoparticles; photoinitiator

from RBCs [10,24–27], the direct interaction between cells and system: camphorquinone (CQ) / amine

RBC surfaces, especially the gene expression of cells on RBCs, and Lucirin (2,4,6-Trimethylbenzoyl-

diphenyl-phosphinoxid; TPO),

requires further study.

phenyl-propanedione (PPD)

The present study examines the DC, the wettability, cell

attachment, cytotoxicity of eluates, and gene expression after

cells were exposed to two different RBCs after they had been a single (blue) emission peak (Translux Wave ; TW, Kulzer

photocured with either a single emission peak or a multiple GmbH, Hanau, Germany) and a dual (violet–blue) emission

emission peak LED LCU. To determine the effect of the vio- peak LED (Translux 2 Wave ; T2W, Kulzer GmbH). Thus, the

let light, a similar amount of blue light was delivered from additional RE from the T2W could be attributed to the violet

both LCUs, and the additional light and RE was provided by the light from this LCU. After they had been made, the specimens

violet LED emitter from one LCU. Human gingival ﬁbroblasts were immediately placed into sterile bags and dark stored at

(HGFs) were cultured (1) directly on the cured RBC materials room temperature.

to evaluate the direct contact interaction between cells and Additional specimens were prepared by Kulzer GmbH:

the materials and (2) in eluates from the materials to assess (1) RBC-V and RBC-P with the surface layer removed and

the effects of the eluted substances on cell behavior. The null polished (surface treated specimens); (2) RBC-V and RBC-P

hypotheses were that the DC, RBC composition, and the addi- without surface treatment; and (3) RBC-V and RBC-P with-

tional violet light from one LCU would not inﬂuence cellular out ﬁller particles (matrix polymer with initiator system

attachment/growth, eluate cytotoxicity, or gene expression. only). These specimens were light cured for 20 s and a dis-

tance of 4 mm using the T2W. The surface treated specimens

were ground with a polishing machine (TegraPol-35, Struers,

␮

2. Materials and methods Ballerup, Denmark). In the ﬁrst step, approximately 50 m

were removed using silicon carbide (P1000/Grit 500). Polishing

2.1. Preparation of test specimens was then performed using silicon carbide (P4000) at 180 rpm.

Both grinding and polishing were performed using ample

Tw o commercial RBCs from the same manufacturer; Venus water cooling.

shade A3 (RBC-V, a submicron hybrid RBC) and Venus Pearl

shade A3 (RBC-P, a nano hybrid RBC) were studied (Table 1). 2.2. Characterization of LCU

The uncured RBCs were ﬁlled into white semi-opaque Teﬂon

rings that had a 5 mm inner diameter and were 2 mm deep. The spectral radiant powers from the two fully charged LCUs

The ﬁlled rings were placed on a glass plate, and the top and were examined in the 350–550 nm wavelength range of the

bottom RBC surfaces covered with a Mylar strip (Patterson, spectrum to determine the REs delivered to the specimens in

Montreal, Quebec, Canada). The light exposure conditions are the blue wavelength range (>420 nm). To measure the power

provided in Table 2. Based on the fact that the manufacturer delivered by the LCUs to the 5 mm diameter specimens, the

recommends a 20 s exposure time for each 2 mm increment of LCUs were ﬁxed in front of a restricted 4 mm aperture into

material, ﬁve different REs were used to irradiate the spec- a 6-inch diameter integrating sphere (Labsphere, North Sut-

imens. These different REs were achieved by adjusting the ton, NH, USA) connected to a ﬁber-optic spectroradiometer

␮

distance between the light tip and the specimen (0, 4 or 12 mm) (USB 4000, Ocean Optics, Dunedin, Fla, USA). A 50 m thick

and by adjusting the exposure time (5, 10 or 20 s). To avoid Mylar Strip (Patterson) was placed over this 4 mm aperture to

any material changes, the specimens were produced under replicate the experimental conditions where the Mylar strip

clean conditions and the operator used gloves and wore a covered the RBC specimen. An internal light source within the

face mask. The specimens were not sterilized or cleaned with sphere (Labsphere) that was traceable to a National Institute

any chemicals, and cell culture experiments did not show of Standards and Technology (Gaithersburg, MD, USA) refer-

any bacterial contamination of these specimens. Every effort ence was used to calibrate the system before measuring the

was made to deliver similar REs of blue-light from both LCUs; LCUs. The lights and exposure conditions were measured in

1176 d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193

Table 2 – Light exposure conditions.

Materials LCU Time % of maximum RE Denomination

5 sec 6.25% V-TW-1

10 sec 12.5% V-TW-2

Translux Wave

20 sec 25% V-TW-3

(TW)

20 sec 50% V-TW-4

® 20 sec 100% V-TW-5

Venus A3 (RBC-V)

5 sec 6.25% V-T2W-1

10 sec 12.5% V-T2W-2

Translux 2 Wave

20 sec 25% V-T2W-3

(T2W)

20 sec 50% V-T2W-4

20 sec 100% V-T2W-5

5 sec 6.25% P-TW-1

10 sec 12.5% P-TW-2

Translux Wave

20 sec 25% P-TW-3

(TW)

20 sec 50% P-TW-4

Venus Pearl A3 20 sec 100% P-TW-5

(RBC-P) 5 sec 6.25% P-T2W-1

10 sec 12.5% P-T2W-2

Translux 2 Wave

20 sec 25% P-T2W-3

(T2W)

20 sec 50% P-T2W-4

20 sec 100% P-T2W-5

triplicate with the LCUs being removed and replaced in front T2W-4 and P-T2W-4). The measurements were performed in

of the sphere for each measurement. triplicate on both sides of each specimen before and after

incubation in a cell culture medium that consisted of DMEM

(Gibco Life Technologies, Carlsbad, CA, USA), 10% fetal bovine

2.3. Degree of conversion (DC)

serum (FBS; Gibco) and 1% antibiotics (Gibco). Each speci-

men was placed in 600 ␮l of cell culture medium for 24 h at

The degree of conversion (DC) was measured using Fourier ◦

37 C without CO2. The contact angles were measured by the

Transform mid-infrared Spectroscopy (Vertex 70, Bruker, Bil-

drop shape analysis system (DAS 10-MK2; Krüss, Hamburg,

lerica, MA, USA). This spectrometer was equipped with a

Germany) using a sessile droplet at room temperature. In sum-

temperature controlled Attenuated Total Reﬂectance (ATR)

mary, a 5 ␮l water droplet from a 0.5 mm diameter tube was

unit (Golden Gate, Specac, Orpington, Kent, UK) containing a

placed on each specimen, and the contact angle was measured

single reﬂection monolithic 2.0 mm × 2.0 mm diamond prism.

using an automatic baseline and circle ﬁtting. The means of

The focus point at the center of the 2 mm × 2 mm square

the left and right contact angles were calculated and used.

diamond crystal is approximately 1 mm in diameter and the

depth of penetration of light into the RBC sample on the ATR

crystal is approximately 2 microns (Specac). The DC of three 2.5. Eluate preparation

replicates of each group was measured at the bottom center

of the specimens. The mid-IR spectrometer data collection Eluates of RBC specimens were prepared following the recom-

started at a rate of 8 measurements per second at a resolu- mendations of ISO 10993-12 using a ratio of 117.8 mm sample

tion of 4 wavenumbers. After 10 s, the LCU was switched on, surface area/ml cell culture medium [28]. Ten cured RBC spec-

and after the DC at the bottom had been recorded for 180 s, a imens taken out from the Teﬂon rings were placed in 6 ml cell

static measurement of the DC was recorded from an average culture medium consisting of DMEM (Gibco Life technologies,

of 20 scans. To calculate the DC, a baseline correction (con- Carlsbad, CA, USA (Gibco)), 10% fetal bovine serum (FBS; Gibco)

◦

cave rubberband) with 5 iterations was performed using the and 1% antibiotics (Gibco) in a 25 ml tube and eluted at 37 C in

Opus v 7.8 software (Bruker, Billerica, MA, USA). The decrease a shaking incubator at 60 rpm (Kuhner AG, Basel, Switzerland).

−1

of the integrated area of an aliphatic C C peak at 1638 cm Eluates were collected every 24 h for 10 days (Fig. 1). 6 ml of cul-

−1 ◦

to an internal reference aromatic C C peak at 1608 cm was ture medium per tube were collected, frozen at −20 C before

used to calculate the DC for RBC-V. RBC-P did not exhibit an cell testing and replaced with 6 ml of additional fresh culture

−1

aromatic peak at 1608 cm that could be used as an inter- medium. The 1st, 2nd, 3rd and 10th eluates were used for the

nal standard. Consequently, the carbon carbon double bond experiments. Control media stored for the same time periods

conversion was calculated from the ratio of the areas of the were used as control eluate samples at each time point.

−1

aliphatic C = C band between 1643–1630 cm wavenum-

bers in the polymerized compared to the unpolymerized RBC.

2.6. Cell culture

2.4. Wettability measurements Primary human gingival tissues were harvested from healthy

donors undergoing third molar extractions, and morpholog-

The wettability was measured on two specimens each for (1) ically spindle-shaped HGFs were cultured as described in

RBC-V cured with T2W for 20 s from a 4-mm distance and previous studies [29,30] after approval of the Ethical Com-

(2) RBC-P cured with T2W for 20 s from a 4 mm distance (V- mittee of the Medical University of Vienna (EK Nr. 631/2007).

d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193 1177

plates. After 24 h of incubation, a cell attachment assay was

performed as described above.

2.8. Cell viability assay with RBC eluates

HGFs were seeded in 96-well plates at a density of 5,000

cells per well. One day after cell seeding, the media were

replaced by eluates collected at each time point. Cell viability

was quantiﬁed, using 3-(4,5-Dimethylthiazol-2-yl)-5(3-

carboxyme-thonyphenol)-2-(4-sulfophenyl)-2H-tetrazolium,

inner salt (MTS) assay (Promega, Madison, WI, USA). This is

a colorimetric method to determine the number of viable

cells based on the principle that mitochondria in viable cells

reduce MTS to a water-soluble coloured formazan product.

The intensity was quantiﬁed using an ELx808 Absorbance

Reader (BIO-TEK, Winooski, VT, USA) after 24 h incuba-

tion/exposure to the eluates. Optical density readings were

converted into percent relative cell viability setting the control

Fig. 1 – Schematic showing how the RBC specimens and

cultures (that contained no toxicant) as 100%.

eluates were used in the experiments.

2.9. Real-time PCR analysis

◦

Cells were cultured in a humidiﬁed atmosphere at 37 C in PCR analyses were performed on cells which had been grown

cell growth medium (DMEM, 10% fetal bovine serum (FBS) and on test materials and cells exposed to eluates from test mate-

1% antibiotics) and detached from tissue culture plastic using rials. For experiments using cells grown on test materials, cells

0.25% EDTA-Trypsin (Gibco) before reaching conﬂuency. The were seeded at a density of 10,000 cells on the materials after

cells used for experimental seeding were from passages 4 to the 3rd elution cycle in 96 well culture plates. For experiments

6. using eluates, 50,000 cells per well in 24 well dishes were

grown for 24 h and then were exposed to the 3rd eluates. Total

2.7. Cell attachment on RBCs RNA was isolated using High Pure RNA Isolation Kit (Roche,

Basel, Switzerland) according to the manufacturer’s instruc-

To determine the cell attachment, nine specimens from each tion either at 3 days post cell seeding on RBCs or 3 days post 3rd

of the 20 groups of RBCs cured under the condition shown in eluate stimulation. A Nanodrop 2000c (Thermo, Wilmington,

Table 2, were exposed three or ten times to cell culture medium DE, USA) was used to quantify total RNA levels. Real-time

with no cells (Fig. 1). Then, cells were seeded onto the RBC discs RT-PCR was performed using a Roche Master mix and quanti-

at a density of 10,000 cells per well in 96-well plates. After 24 h ﬁed on an Applied Biosystems 7500 Real-Time PCR Machine.

◦

of incubation in a humid atmosphere of 5% CO2 at 37 C, the Primer sequences for genes encoding interleukin-1, -6, -8 and

␣

cells were ﬁxed with 4% formaldehyde for 2 min. Then cells -11 (IL-1, IL-6, IL-8 and IL-11), tumor necrosis factor- (TNF-

␣

␤ ␤ ␤

were permeabilized by methanol for 20 min and stained with ), integrin 3, ki-67, Transforming growth factor (TGF- ),

Giemsa solution (MERCK, Darmstadt, Germany) for 15 min. platelet-derived growth factor (PDGF), adrenomedullin (ADM),

␤

After rinsing in sterile water and drying, the specimens were caspase 3, 8 and 9, annexin A5, and -actin were fabricated

examined under a microscope (Olympus SZX7, Tokyo, Japan) with primer sequences according to Table 3. The Ct method

and the numbers of cells on the RBCs were counted. One image (a comparative CT method) was used to calculate gene expres-

per sample was taken at 5.6× magniﬁcation from the middle sion levels normalized to the expression of the housekeeping

␤

portion of the sample. Cells were distinguished by their shape gene -actin.

as either round or spread cells and counted manually by using

the counting tool in Photoshop CS5.1 software (Adobe, San 2.10. Protein release quantiﬁcation (ELISA)

Jose, CA, USA).

The additional groups of RBCs (1) RBC-V and RBC-P with To determine the quantity of cytokine release from HGFs

surface treatment, (2) RBC-V and RBC-P without surface treat- after exposure to RBCs or RBC eluates, ELISA quantiﬁcation

ment, and (3) RBC-V and RBC-P without ﬁller particles were assays were utilized. For experiments using cells grown on

further tested after the 10th elution, and cell attachment was test RBCs, cells were seeded at a density of 10,000 cells on

recorded after 24 h of incubation. the materials after the 3rd elution cycle in 96 well culture

Cell attachment on either the top or bottom of the RBCs was plates. For experiments using eluates, cells were cultured

compared in three groups; of V-T2W-1, V-T2W-4, and V-T2W-5 at a density of 50,000 cells per well in 24 well dishes and

(see Table 2) and specimens were prepared and marked with exposed to 3rd eluates. The supernatants were collected after

a scratch on the top surface of the composite. 10 specimens 3 days incubation for both experiments of culture on RBC

per test group were used just after the 3rd eluate collection. and in RBC eluates. IL-6 (DY206-05, range = 9.38–600 pg/mL),

Cells were seeded on either the top or bottom surface of the TGF-␤ (DY240, range = 31.20–2000 pg/ml) and VEGF (DY293B-

composite discs at a density of 10,000 cells per well in 96-well 05, range = 31.20–2000 pg/mL) were quantiﬁed using an ELISA

1178 d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193

Table 3 – List of primer sequences for real-time PCR analysis.

Gene Forward primer Reverse primer

hIL-1 ggttgagtttaagccaatcca ggtgatgacctaggcttgatg

hIL-6 gaaaggagacatgtaacaagagt gattttcaccaggcaagtct

hIL-8 atgacttccaagctggccgt tccttggcaaaactgcacct

hTNF-␣ cagcctcttctccttcctgat gccagagggctgattagaga

hIntegrin ␤3 gtgctgacgctaactgacc catggtagtggaggcagagt

hKi-67 gaggtgtgcagaaaatccaaa ctgtccctatgacttctggttgt

hTGF-␤ actactacgccaaggaggtcac tgcttgaacttgtcatagatttcg

hPDGF cacacctcctcgctgtagtattta gttatcggtgtaaatgtcatccaa

hADM ggacatgaagggtgcctctc tgttcatgctctggcggtag

hIL-11 tgcacctgacacttgactgg agtcttcagcagcagcagtc

hCaspase 3 ttcagaggggatcgttgtagaagtc caagcttgtcggcatactgtttcag

hCaspase 8 ctgctggggatggccactgtg tcgcctcgaggacatcgctctc

hCaspase 9 atggacgaagcggatcggcggctcc gcaccactgggggtaaggttttctag

hAnnexin A5 cagtctaggtgcagctgccg ggtgaagcaggaccagactgt

h␤-actin ccaaccgcgagaagatga ccagaggcgtacagggatag

kit (DuoSet, R&D Systems, Minneapolis, MN, USA) according

3. Results

to the manufacturer’s protocol. Absorbance was measured at

450 nm and 570 nm using an ELx808 Absorbance Reader, and

3.1. Characterization of the LCUs

OD values at 570 nm were subtracted from the readings at

450 nm (OD450 – OD570). Each protein concentration was cal-

Representative emission spectra from the two LCUs are shown

culated from the standard carve.

in Fig. 2. They clearly show the single emission peak for TW in

the blue wavelength area and the two peaks for the T2W with

the additional peak in the violet wavelength region. However,

by adjusting the distance from the target, the peaks in the blue

2.11. Statistical analysis

region above 420 nm became similar for both LCUs.

As shown in Table 4, the REs (J/cm ) delivered to the spec-

To characterize the LCUs, three separate measurements 2 2

imens ranged from 1.5 J/cm from the TW and 1.7 J/cm from

were made of the spectral radiant power and the radi- 2 2

the T2W to 24.5 J/cm from TW and 28.9. J/cm from the

ant power emitted between 350 nm and 550 nm and the

T2W. The peak wavelength in the blue spectrum from both

mean values with the standard deviation (SD) are reported

lights was almost identical (see above) with only differences

(Table 4).

in the violet wavelength range. Although both lights delivered

To determine the DC for the RBCs made using the dif-

the same power and RE in the blue range (>420 nm), when

ferent radiant exposures, three independent specimens were

the violet light was included, the maximum RE delivered to

tested per group. The DC data for each composite was ana- 2

the specimens increased to 28.9 J/cm from the T2W. Thus,

lyzed separately using a two-way ANOVA on the raw data and

approximately 15% of the light from the T2W is from the vio-

the normalized data using the Box-Cox transformation. Mean

let LED, and this is the reason why the RE is greater from the

DC results with SD are reported in Table 5.

T2W compared to TW.

To determine the wettability, six independent measure-

ments were performed. The medians with 25/75 percentiles

are reported, and statistical evaluation was performed using 3.2. Degree of conversion (DC)

a non-parametric Kruskal–Wallis test and Wilcoxon matched-

pairs rank test at p < 0.05. The results for the degree of conversion (DC) are reported in

For cell attachment assay on RBCs, nine independent spec- Table 5 and Fig. 3. Both materials (RBC-V and RBC-P) demon-

imens were assessed. Cell proliferation assays testing eluates strated a logarithmic relationship (R ranging from 0.944 to

were repeated 3 times in triplicate. Experiments for real- 0.973) between the RE and the DC achieved (Fig. 3). The mea-

time PCR and protein release assays were repeated 3 times sured DC of the RBC-P was generally lower compared to RBC-V.

in duplicate. Medians and interquartile ranges are reported A plateau for DC was reached after a RE of about 15 J/cm ,

in 5–7,10–12 and 15. For the real time PCR medians of the except for the combination of RBC-P and TW (about 25 J/cm ).

three independent experiment are reported (8, 10, 13, 14). The results for the raw and the normalized data agreed. Both

The values were analyzed for statistical signiﬁcance using factors (LCU and RE) and their interaction were signiﬁcant

the non-parametric Kruskal–Wallis test for cell proliferation (p < 0.001).

assay, gene expression analysis, and protein release assay Using T2W matching the RE in the blue wavelength range to

using GraphPad Prism 6.0 software (GraphPad Software, Inc., TW exposed for 10 s and using the 25% level as the reference,

La Jolla, CA, USA). For all statistical tests, the signiﬁcance level in both cases, and for both RBCs, as the RE increased, the DC

of p values was set to <0.05. Correlation analyses relating DC increased, but the increase in DC was the greatest between the

to REs or cell reaction to DC were performed using the original two lowest REs.

data, and each regression curve was calculated using Graph- For RBC-V, the T2W always produced the greatest DC values

Pad Prism 6.0 software. (p = 0.007). For RBC-P, the T2W produced the greatest DC values

d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193 1179

Table 4 – Light exposure conditions and radiant exposure (RE) for the test specimens; wavelength range 350–550 nm;

radiant exposure for T2W refers to the radiant exposure of the combined emissions in the blue and the violet regions

(means ± SD, n = 3). The additional radiant exposure comes from the violet range of the T2W.

Irradiance Radiant exposure (RE) (J/cm ± SD)

LCU % Power 2

(mW/cm ± SD)

Time 5 (sec) Time 10 (sec) Time 20 (sec)

100 1227 ± 20 24.5 ± 0.4

50 622 ± 28 12.4 ± 0.6

(350–550 nm)

25 309 ± 6 1.5 ± 0.1 3.1 ± 0.1 6.2 ± 0.1

± ±

100 1143 23 28.9 0.5

T2W Matching the Blue

± ±

50 639 18 13.9 0.4

Spectrum to the TW

25 344 31 1.7 ± 0.2 3.4 ± 0.3 6.9 ± 0.6

Table 5 – Radiant exposures, REs and % degree of conversion, DC (means and SD, n = 3).

Materials LCU Radiant exposure (J/cm ) DC (%) SD

1.5 25.2 2.1

3.1 40.4 1.9

TW 6.2 45.4 1.6

12.4 47.6 1.9

24.5 50.3 1.7

RBC-V

T2W - Matching the RE 1.7 25.6 2.2

in the blue range with 3.4 41.3 2.3

the additional energy 6.9 46.3 0.8

coming from the violet 13.9 49.4 0.8

range 28.9 51.9 0.7

1.5 0.7 1

3.1 27.8 1.8

TW 6.2 36 3.4

12.4 40.6 2.6

24.5 48.8 1

RBC-P

T2W - Matching the RE 1.7 0.7 1

in the blue range with 3.4 30.1 2.5

the additional energy 6.9 39.2 1.7

coming from the violet 13.9 44.9 2.1

range 28.9 48.2 1.7

Fig. 2 – Radiant power delivered to the specimens from the Translux Wave (TW) and Translux 2 Wave (T2W); both LCUs

delivered similar powers (Pblue) above 420 nm, but the T2W delivered additional power (Pviolet) below 420 nm.

◦ ◦ ◦

(p = 0.0001) in all cases except for at the highest RE level where 4) ranged between 68.4 and 81.5 for RBC-V and between 67.4

◦

they were not signiﬁcantly different (p = 0.139). and 90.4 for RBC-P. After 24-h incubation in the cell culture

medium, the contact angles on both, RBC-V and RBC-P, sig-

niﬁcantly decreased when compared to initial values (Fig. 4).

3.3. Wettability

However, there were no signiﬁcant differences in wettability

between the two RBCs (V-T2W-4 and P-T2W-4).

Before elution in the cell culture medium, contact angles for

both RBC-V and RBC-P cured under the same conditions (T2W-

1180 d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193

Fig. 3 – Degree of Conversion (DC) on RBC-V and RBC-P

related to the radiant exposure RE delivered by the TW and

T2W LCUs. The graph represents the medians with 25/75

percentiles and ﬁtting curves (one-phase association line,

least squares ﬁt).

Fig. 5 – Relative cell attachment after 24 h when seeded

either on the top or bottom surface of 3 different composites

(V-T2W-1, V-T2W-4, and V-T2W-5) after 3 elution cycles.

Both the numbers of round cells and spreading cells

attached to the specimens were calculated. No signiﬁcant

differences were observed between the top and bottom of

the specimens. (100% = optimal density reading of control

cultures. p < 0.05. n = 9 medians with 25/75 percentiles).

cured from one side (top) did not result in a signiﬁcantly dif-

ferent cell growth rate between the top and the bottom on the

RBC specimens for the whole range of REs used in this study.

Therefore, the cell attachment and growth were compared for

all 20 groups of RBCs without the distinction of the sides (top

and bottom) of specimens after 3 and 10 elution cycles in the

cell culture medium.

Fig. 6 shows that RBC-V specimens generally had higher

cell numbers when compared to RBC-P and cells showed a

Fig. 4 – Changes of contact angles of the top and bottom rounder morphology (indicating some alteration) on RBC-P,

surfaces of composites before and after elution in cell but less or none was observed on samples of RBC-V. This was

culture medium for 24 h. (* denotes a signiﬁcant difference observed for the majority of specimens both after 3 and 10

between groups, p < 0.05, ** denotes signiﬁcantly higher elution cycles. Cell attachment was consistently improved for

than all other initial specimens before elution, p < 0.05, n = 6 the 10th elution when compared to the 3rd elution, which was

medians with 25/75 percentiles). statistically signiﬁcant for REs of 6.9 and 13.9 J/cm for V-T2W,

2 2 2

>6.2 J/cm for V-TW, >6.9 J/cm for P-T2W, and >12.4 J/cm for

P-TW. Moreover, in both RBC-V and RBC-P specimens, more

3.4. Cell attachment and growth on RBCs viable cells were observed on the RBCs photo-cured with T2W

when compared to RBCs photo-cured with TW. The lowest

Preliminary experiments showed that cell growth on the number of cells were observed when RBC-P was photocured

composite surface (data not shown) did not occur on speci- with the single peak TW LCU.

mens until after one or two elution cycles in cell-free culture When the surface-treated RBC specimens and the speci-

medium. Therefore, the tests were performed on the RBCs mens containing the matrix components only with no ﬁller

after three and ten elutions in the cell culture medium. The were investigated for cell attachment (Supp. Fig. 4), the surface

surfaces of the two sides of the specimens were compared treatment (removal of surface layer) of RBC-V did not substan-

for cell attachment and growth to assess the effect of sample tially inﬂuence cell attachment. In contrast, surface treatment

thickness (Fig. 5, Supp. Fig. 1). These tests were performed on on RBC-P improvement the cell attachment (Supp. Fig. 4a). The

RBC-V after the 3 elution cycles of the specimens cured with 3 matrix only specimens of RBC-V showed good cell attachment

different REs (1.7, 13.9 and 28.9 J/cm ) from the T2W LCU. There whereas no cell attachment was observed on the matrix only

was no signiﬁcant difference for cell attachment between the specimens of RBC-P (Supp. Fig. 4b).

top and bottom of specimens in all groups (Fig. 5, Supp. Fig. The inﬂuence of the different REs and the respective DCs

1). These data suggest that the 2 mm thickness of specimens upon cell numbers (both spread and round cells) on RBC spec-

d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193 1181

Fig. 6 – Relative cell attachment on RBCs at 24 h after 3 elution cycles and 24 h after 10 elution cycles. Both the numbers of

round cells and spreading cells were calculated. (100% = optimal density reading of control cultures. #; signiﬁcantly lower

than the plastic control samples, p < 0.05. n = 9 medians with 25/75 percentiles).

Fig. 7 – Inﬂuence of (a, b) radiant exposure and (c, d) DC on cell attachment (relative total cell number including both round

and spreading cells) on composites after (a, c) 3 and (b, d) elution cycles at 24 h. The graphs represent (a, b) the medians

with error bars 25/75 percentiles and ﬁtting curves (semi-log line; x is slog, y is linear, least squares ﬁt), and (c, d) median

and ﬁtting curves (log-log line; least squares ﬁt) of the 9 specimens’ values. (100% = optimal density reading of cultures on

the plastic control).

imens is depicted in Fig. 7. Again, the above mentioned lower were counted on T2W cured specimens compared to those

number of cells on RBC-P compared to RBC-V become appar- cured with TW. The combination of RBC-V with TW yielded

ent when plotting the cell numbers vs. the different REs (Fig. 7a low cell numbers even at the highest RE. The curve ﬁtting

and b) especially for the 3rd elution. Furthermore, more cells resulted in a logarithmic equation for both specimens after

1182 d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193

Fig. 8 – Results from real-time PCR of HGFs seeded on RBCs at 3 days for genes encoding (a) IL-1, (b) IL-6, (c) IL-8, (d) IL-11, (e)

␤

TGF- , (f) TNF-␣. (* denotes signiﬁcantly higher than the plastic control, p < 0.05, n = 3, medians with min and max values ◦

represented by ).

3rd and 10th elution cycles. A plateau was reached for RBC-V inﬂammatory cytokines (IL-1, IL-6, IL-8, and IL-11) on RBC-V

at about 15–20 J/cm for the 10th elution time point. For the specimens (Fig. 8). Also, specimens cured with T2W in most

3rd elution, no plateau was observed. cases – and especially with IL-6 – evoked a smaller increase

Interestingly, plotting the cell numbers vs. the DC (Fig. 7c of the gene expression than specimens cured with TW. Fur-

and d), the above-seen differences (Fig. 7a and b) were not or thermore, mRNA levels of ki-67 (a marker for cell proliferation)

they were less apparent. For both, the 3rd and the 10th elu- were signiﬁcantly downregulated on the less cured specimens

tion time point a logarithmic relation between cell numbers when compared to control tissue culture plastic, especially

and DC can be observed. At a DC of around 50%, the cell num- for the combination of RBC-V and TW. Better photo-curing

ber tended to reach its maximum at the 10th elution period increased mRNA levels of the proliferation marker (ki-67) on

(except for RBC-P with TW). This can be considered to be the RBC-V specimens cured by the TW LCU (Fig. 9). Also, mRNA of

DC which should be reached for optimal cell growth on the apoptose markers (caspase 9, annexin A5) were upregulated

respective specimens. However, this value was not reached at at low REs, especially for RBC-V. The expression of apoptose

the 3rd elution period. markers Caspase 3 and 8 were upregulated, but not statisti-

cally signiﬁcant. Parallel to this, the gene expression for ADM

(adrenomedullin), a marker for increasing cell tolerance to

3.5. Gene expression of cells grown on RBCs

oxidative stress, was signiﬁcantly upregulated when a low

RE was delivered to RBC-V. Similarly, the IL-6 released from

Based on the results of cell growth potentials on RBCs, gene

cells seeded on RBC-V was signiﬁcantly upregulated when

expression and protein release from HGFs cultured for 3 days

compared to control tissue culture plastic (Fig. 10). Of note,

were recorded for RBC-V cured with both LCUs at two different

better photo-curing tended to decrease the release of IL-6 from

REs (20 s 25% of maximum exposure and 20 s 50% of maximum cells.

exposure) (Figs. 8 and 9). Respective tests with RBC-P could

In summary, the results of gene expression of HGF on RBC

not be performed due to the low cell numbers on this mate-

specimens demonstrated that the lower DC/RE and using the

rial. The mRNA levels of IL-1, -6, -8, -11, TGF-␤, integrin ␤3,

TW rather than T2W LCU up-regulated inﬂammation markers

caspase-9, annexin A5, and ADM on V-TW-3 specimens were

level and apoptosis markers level and decreased the prolifer-

signiﬁcantly upregulated when compared to control on the tis-

ation marker, ki-67. Similarly, the results of protein synthesis

sue culture plastic. Interestingly, better photo-curing (20 s, 50%

of maximum exposure) generally decreased mRNA levels of

d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193 1183

Fig. 9 – Results from real-time PCR of HGFs seeded on RBCs after 3 Elution cycles at 3 days for genes encoding (a) ki-67, (b)

integrin ␤3, (c) caspase 3, (d) caspase 8, (e) caspase 9, (f) annexin A5, (g) ADM, (h) PDGF. (* denotes signiﬁcantly higher than

the plastic control, p < 0.05, # denotes signiﬁcantly lower than the plastic control, p < 0.05, n = 3, medians with min and max

◦

values represented by ).

of HGF on RBC showed that at lower DC/REs using the TW LCU viability when compared to the control medium anymore.

tended to increase the inﬂammation markers level. Thus, differences between the groups became more appar-

ent at low numbers of elution cycles. Furthermore, the groups

receiving less RE showed signiﬁcantly lower cell viability in

3.6. Cell viability after eluate exposure

the 1st, 2nd, and 3rd eluates. In detail, eluates from RBC-V

of the 1st, 2nd and 3rd eluates) (with few exceptions) evoked

Cytotoxicity of RBC eluates collected at 4 different time points

lower cell viability when compared RBC-P (Fig. 11a–c). For

(1st, 2nd, 3rd, and 10th eluate) of the 20 groups after 24 h incu-

the ﬁrst eluate, the combination of RBC-P and the TW LCU

bation/exposure are shown in Fig. 11. In general, the eluates

resulted in lower cell viability than using the same material

collected at the later time points demonstrated less cytotox-

with T2W.

icity for both, RBC-V and RBC-P, as well as for both LCUs.

The eluates of the 10th elution did not negatively affect cell

1184 d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193

The inﬂuence of the different REs and the respective DCs

on eluate cell viability for the 1st and the 3rd eluate is shown

in Fig. 12. For the 10th eluate, no such inﬂuence was appar-

ent from data presented in Fig. 11. The plots of cell viability

vs. REs (Fig. 11a–c) show a logarithmic shape with an ini-

tially steep increase of cell viability with increasing RE, and

a plateau is reached between 5 and 10 J/cm . Furthermore, the

ﬁrst and second eluates from RBC-V cured with both LCUs

evoked mainly less cell viability than the eluates from RBC-

P, e.g., for RE of >3.4 J/cm from T2W LCU when cultured in the

ﬁrst eluate.

Comparison of the effect of the two LCUs revealed lower cell

viability of the RBC specimens photocured with TW compared

to the T2W LCU when testing the ﬁrst eluate. For the ﬁrst and

the second eluate, the combination of RBC-V with TW yielded

the lowest cell viability.

The inﬂuence of the different DCs upon cell viability

(Fig. 12d–e) also shows a logarithmic curve shape for RBC-P

for all eluates. For RBC-V, which generally showed lower cell

viability at the same DCs as RBC-P, the data point for low

DCs could not be reached (despite low RE) due to the exper-

imental setup. Again, a logarithmic shape of the curves is

Fig. 10 – Protein release from HGFs seeded on composites apparent. Here, the same basic characteristics as described

after 3 elution cycles at 3 days of IL-6. (* denotes above (Fig. 12a–c) can be observed, here even at the 3rd eluate.

signiﬁcantly higher than control plastic, p < 0.05, p < 0.05, For the 1st elution, 100% cell viability indicating the DC,

n = 6, medians with 25/75 percentiles). which does not produce eluate cytotoxicity, was only found for

RBC-P together with T2W at a DC of around 40% (Fig. 12d–f). For

the 2nd elution, it was around 30% for RBC-P and 50% for RBC-

Fig. 11 – Relative cell viability of HGFs stimulated in eluates of composite materials at 24 h in (a) 1st eluates, (b) 2nd eluates,

(c) 3rd eluates, and (d) 10th eluates. (* denotes signiﬁcantly higher than control plastic, p < 0.05, # denotes signiﬁcantly lower

than the plastic control, p < 0.05, n = 9, medians with 25/75 percentiles).

d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193 1185

Fig. 12 – Inﬂuence of (a–c) radiant exposure and (d–f) DC on cell viability in (a, d) 1st eluates, (b, e) 2nd eluates, (c, f) 3rd

eluates at 24 h. The graphs represent (a–c) the medians with error bars 25/75 percentiles and ﬁtting curves (one-phase

association line, least squares fit), and (d–f) median and fitting line (log-log line; least squares fit) of the 9 specimens’ values.

(100% = cell viability in control media).

V. For the 3rd elution the differences between the materials Protein release patterns for measured cell markers from

further decreased (Fig. 12d–f). cells cultured in eluates were not consistent with mRNA lev-

els of those cells (Fig. 15). Especially in RBC-P groups, using a

low RE signiﬁcantly decreased released protein levels of IL-6,

3.7. Gene expression of cells after eluate exposure ␤

TGF- , and VEGF (Fig. 15a–d). Interestingly, eluates from RBC-V

especially at low REs increased the expression of Il-6, and the

Lastly, the gene expressions and protein release from HGFs

inﬂuence of the RE upon the biological reaction can be seen.

exposed to 3rd RBC eluates were investigated after 3 days

IL-1, TNF-␣, PDGF-BB release was not detected by ELISA (data

incubation/exposure (Figs. 13–15). The mRNA levels of inﬂam-

not shown).

mation mediators IL-1, -8, and TNF-␣ from the least cured

In summary, the results of gene expression of HGF in elu-

RBC-V, and RBC-P specimens by TW, (but not by T2W) were

ates demonstrated that at the lower REs and DC, using the

signiﬁcantly upregulated when compared with control tissue

TW LCU rather than T2W up-regulated inﬂammation and

culture material (Fig. 13a–c, f). The expression of mRNA for the

apoptosis markers levels and decreased the expression of

inﬂammation mediator Il-6 was signiﬁcantly upregulated for

mRNA of the proliferation marker, ki-67. Both RBC-V and RBC-

the RBC-P in combination with TW, for RBC-V with TW upreg-

P showed similar inﬂammation marker expression pattern,

ulation was not statistically signiﬁcant. For IL-11, a tendency

whereas RBC-P, especially the combination of RBC-P and TW

of upregulation of mRNA after exposure to eluates from the

showed higher ki-67, a proliferation marker when compared

least cured RBC-V specimen with TW could be observed, but

to the RBC-V. The results of protein synthesis of HGF in elu-

this was not statistically signiﬁcant (Fig. 13d).

ates showed that at the lower REs and DC using RBC-P rather

However, mRNA levels of PDGF (Fig. 14h) were signiﬁcantly

than RBC-V increased the inﬂammation and the angiogenesis

downregulated under the least well-cured conditions in RBC-P

marker levels.

specimens cured with either LCUs when compared to control

medium. In most specimens receiving higher RE, a decrease

in mRNA expressions of inﬂammatory cytokines especially in 4. Discussion

groups treated with TW was observed. Nevertheless, no signif-

icant differences were found in any groups in apoptotic gene The present study aimed to investigate the inﬂuence of a

levels, including caspase 3, 8, 9 and annexin A5 (Fig. 14c–f). range of radiant exposures and respective degrees of conver-

1186 d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193

Fig. 13 – Results of real-time PCR of HGFs stimulated in composite eluates at 3 days for genes encoding (a) IL-1, (b) IL-6, (c)

IL-8, (d) IL-11, (e) TGF-␤, (f) TNF-␣ (* denotes signiﬁcantly higher than the plastic control, # denotes signiﬁcantly lower than

◦

the plastic control, p < 0.05, n = 3, medians with min and max values represented by ).

sion of two commercially available RBCs upon the reaction points. The RBCs were speciﬁcally chosen because, according

of gingival ﬁbroblasts (HGFs), which were in direct con- to the information from the manufacturer, they differ in the

tact with the test materials or exposed to material eluates. chemical composition of the polymer matrix, one being Bis-

Tests were performed under different experimental con- GMA- based (RBC-V), the second being based on UDMA (RBC-P)

ditions using RBCs with different compositions and two (Table 1). Furthermore, both materials differed in the initiator

different LED-based LCUs, one emitting light above 420 nm systems. According to the information from the manufac-

(mono wave LCU), the other emitting an additional violet turer, the conventional initiator system of RBC-V is based on

peak (a dual emission peak LCU). The intent was to deliver CQ/Amine, whereas RBC-P is based on CQ with a blend of violet

similar amounts of blue (>420 nm) energy to the RBCs so sensitive co-initiators. The inhibitor is the same in both mate-

that the potential beneﬁts of the violet light present could rials. The LCUs were from the same manufacturer and were

be determined. Finally, the inﬂuence of elution on mate- selected according to the different initiator systems used in

rial cytotoxicity was evaluated simulating a more prolonged two RBCs of this study, where especially RBC-P was intended

exposure. to be cured with a dual emission peak LCU that delivers violet

Particular emphasis in this study was placed on testing light as well as blue light.

strictly characterized RBC specimens with deﬁned surfaces HGFs were used as target cells as it is well recognized that

in different cell culture systems and using different end- the primary cells to contact RBCs clinically are gingival epithe-

d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193 1187

Fig. 14 – Results of real-time PCR of HGFs stimulated in 3rd composite eluates at 3 days for genes encoding (a) ki-67, (b)

integrin ␤3, (c) caspase 3, (d) caspase 8, (e) caspase 9, (f) annexin A5, (g) ADM, (h) PDGF. (# denotes signiﬁcantly lower than

◦

the plastic control, p < 0.05, n = 3, medians with min and max values represented by ).

1188 d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193

Fig. 15 – Protein release from HGFs seeded in 3rd composites eluates at 3 days of (a) IL-6, (b) TGF-␤, (c) VEGF. (# denotes

signiﬁcantly lower than the plastic control, p < 0.05, n = 6, medians with 25/75 percentiles).

lial cells [31]. When preparing deep cavities reaching into the RBC-P. These results are in agreement with other reports. A

gingival sulcus, the gingival epithelium is inevitably touched higher DC of a Bis-GMA-based RBC compared to a urethane

and potentially removed. Thus, the newly placed RBC may also based RBC at the same REs was also found in a recent study,

contact gingival ﬁbroblasts, which also play an essential role in where RBC specimens had been exposed to either 5 or 10 J/cm

the resulting healing process. Furthermore, HGFs show better [32] where it was also reported that the DC of a Bis-GMA-

2 2

growth kinetics and ease of use compared to human gingival based RBC was lower after 5 J/cm than after 10 J/cm RE. A

epithelial cells. logarithmic relationship between DC and exposure time was

Cell viability assays with eluates were performed accord- also reported by Cardoso et al. [6]. Another study directly mea-

ing to ISO standards (ISO 10993-5 and 10993-12) as previously sured the DC of RBCs after REs ranging from 3 to 48 J/cm ,

reported [21], to complement the experiments for cell adhe- which comes close to the range of the present study [33].

sion. In order to provide more mechanistic background Again, a logarithmic relationship was found between the DC

information, gene expression of HGFs was recorded for rele- and RE. In the present study, the presence of the violet peak

vant genetic markers for inﬂammation IL-1, IL-6, IL-8, IL-11, in the T2W LCU resulted in a small, but statistically signiﬁ-

and TNF-␣), apoptosis (caspase 3, caspase 8, caspase 9, cant, increase in DC even when similar REs were delivered.

annexin A5) or regenerative processes (ki-67, integrin ␤3, ADM, The most likely explanation is that the CQ initiator is activated

PDGF and TGF-␤). To the best of our knowledge, such combined by blue light from the dual emission peak LCU, and the addi-

experiments with well-deﬁned test specimens have not been tion of Lucirin TPO initiator that has an optimal absorbance

done previously. at around 385 nm and phenyl-propanedione that has an opti-

mal absorbance at around 398 nm, to RBC-P, but not in RBC-V,

4.1. Radiant exposures and resulting DC improved the polymerization. Since the TW LCU delivered no

light in the lower violet wavelength, this LCU produced the

The REs delivered to the specimens ranged from 1.5 J/cm to lowest DCs (Table 5). This observation reinforces the need to

28.9 J/cm and were selected based on clinical considerations match the emission spectrum from the LCU with the absorp-

for different distances from the tip of the light guide to the tion spectrum of the photoinitiators within the RBC. In the

material (0–12 mm) and different exposure times (5–20 s). We study mentioned above a plateau for DC was observed after

found a logarithmic relationship between REs and DC and a delivering a RE of 20–25 J/cm , which is in the range of in the

2 2

generally higher DC at the same REs for RBC-V compared to present results being around 15 J/cm for RBC-V and 25 J/cm

d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193 1189

for RBC-P cured with the single (blue light) emission peak LCU 4.4. Materials composition

[33].

Cell attachment was reduced when in direct contact with RBC-

P in comparison to RBC-V (Figs. 6 and 7, and Supp, Figs. 2

and 3). This difference cannot be explained by differences

4.2. Degree of conversion and cell reactions

in wettability between the two RBCs, because contact angle

measurements showed no differences between both RBCs

In the present study, virtually all the observed cell reactions

(Fig. 4). The change of wettability after incubation with cell cul-

were dependent upon the degree of conversion (DC) that, in

ture medium containing serum is likely due to the formation

turn, was dependent upon the RE, the material composition

of a proteinous coating [39]. Also, eluate cytotoxicity cannot

and the compatibility between the photoinitiators and the

explain this difference, because opposing results were found

emission spectrum from the LCU. It is known from the lit-

(see also below).

erature that the higher the DC of tested RBCs, the lower is

However, the composition of the resin matrix in the RBC

the amount of substances that are released [13,24,34] and this

materials was different. Specimens, where the surface layer

explains the observation that lower DC values result in less

had been removed, showed a substantially increased cell

cell attachment and higher eluate cytotoxicity. Of note, we

attachment for RBC-P, whereas for RBC-V it stayed at the same

were able to parallel the differences in DC for each material

high level (Supp. Fig. 4a). The surface layer of RBCs may poten-

with data from the biological experiments. For instance, the

tially be more resin rich, and if this is removed by polishing,

urethane-based RBC-P showed a lower DC compared to the

the cells are exposed to less resin and more ﬁller particles [40].

Bis-GMA based RBC-V and thus lower cell numbers on RBC-

Furthermore, specimens made by Kulzer GmbH that included

P were observed than on RBC-V. Especially the combination

just the polymer matrix did not show any cell attachment for

of the RBC containing two initiator systems (RBC-P) with the

RBC-P, but did show good attachment for RBC-V (Supp. Fig.

LCU, which only referred to one initiator (camphorquinone)

4b). From these results, it appears that the difference in cell

showed consistently lower DC on the one side and lower

attachment observed between the two RBCs from the same

cell attachment, higher eluate cytotoxicity or higher expres-

manufacturer (RBC-V and RBC-P) was due to their different

sion of inﬂammation or apoptosis markers on the other side.

chemical compositions.

This demonstrates that the biological system very sensitively

Interestingly the difference in cell attachment between the

reacts towards differences in DC, at least shortly after photo-

two RBCs was not or less apparent when the cell numbers were

curing. Interestingly, no such relationship could be observed

plotted against the DC (Fig. 5). Therefore, it can be assumed

between the lower DC of RBC-P and the eluate cytotoxic-

that the different DC characteristics may also play a role in

ity, which proved to be lower than RBC-V. This indicates

the differences in cell attachments behavior of the two RBCs.

that a more detailed and quantitative analysis of the rela-

Furthermore, it has been reported that cell attachment and

tionship between DC and the cellular response should be

proliferation depend on surface chemistry and varies with

conducted.

individual functional groups rather than general surface prop-

erties such as wettability [20].

4.3. Elution cycles

4.5. Biological endpoints

In the present study, increasing the number of elution cycles

The difference mentioned above in cell attachment between

lead to more cell attachment and less eluate cytotoxicity. This

the two RBCs was contrary to the results of the eluate cell

corroborates the results from previous studies that reported

viability tests, where RBC-P was less cytotoxic at correspond-

decreased elution of substances with increasing elution time

ing REs and DCs especially for the 1st and 2nd eluate than

[35] or with repeated elutions [12,36]. Also, after several elu-

RBC-V (Figs. 11 and 12). The cytotoxicity of eluates depends on

tions, the material surfaces may be covered with a protein

the nature and the effective quantity/concentration of eluted

layer from the cell culture medium, which also may favor

substances, e.g., monomers [34,41,42]. In cultures of human

cell attachment [37]. Repeated elution cycles might affect DC.

gingival ﬁbroblasts, a UDMA monomer (a component of RBC-P)

However, in a recent study, most of the post-curing was found

proved to be less cytotoxic than a Bis-GMA monomer (a com-

only up to 15 h after light exposure [38]. In the present study,

ponent of RBC-V) [34]. In the present study, the eluates were

the DC measurement was performed at 180 s after the LCU

not further analyzed as this was beyond the scope of the study.

was turned on and likely would have increased during storage

However, this difference between the two monomers may be

due to the effects of post-curing. However, the elution cycles

a reason for the lower cytotoxicity of RBC-P eluate compared

took place much longer than 15 h after light exposure. Thus,

to RBC-V. On the other hand, this reason cannot explain the

the effects of post-curing can be ruled out as a reason for the

difference in cell reaction in direct contact with the test mate-

decreasing toxicity after multiple elution cycles. Furthermore,

rials, which was the opposite compared to the elution test. It

the leachable components within the RBCs will decrease after

can be speculated that either differences in surface chemistry

each elution. However, during the early phase of the material

(see above) or local concentration effects are responsible for

application, healing processes, e.g. of damaged pulpal or gin-

this reaction. Time-dependent concentrations of substances

gival tissues may take place, and this timeframe is reﬂected

released directly onto contacting cells may be different from

in the results from tests using the eluates from the ﬁrst few

a situation, where substances ﬁrst are release into cell cul-

elution cycles (see below).

ture medium over some time (e.g., 24 h) and then exposed to

1190 d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193

cells due to dilution or possible interactions with the culture of gingival tissues in contact with RBCs that have a low DC

medium. This is in line with the results of this study showing may be inhibited.

a more sensitive gene expression pattern for cell attachment TGF-ˇ was signiﬁcantly increased when cells were in direct

assays than for elution tests. contact with RBC that had a low DC. TGF-␤ has many func-

Despite the difference between the two materials, both, cell tions in the oral environment; e.g., stimulating tertiary dentin

attachment assay and the eluate assay demonstrated higher formation [51]. It is also involved in wound healing: in experi-

cell toxicity on the specimens with fewer elution cycles, and mental animals, TGF-␤ signal transduction blockage increased

lower DC again showing the interrelation of the factors. For healing [52,53]. However, it also plays a role in keeping the

biological characterization of a restorative material, however, ROS-balance of the cells by increasing the uptake of cystine

both the direct contact situation with relevant target cells leading to better maintenance of cellular glutathione during

and standard elution tests should be considered because the oxidative stress [54]. RBC monomers such as TEGDMA and

results may differ between these two test methods and the HEMA have been shown to cause a ROS imbalance [55,56] and

direct cell situation better reﬂects the local oral conditions in the upregulation of TGF-␤ can be regarded as an adaptive cell

the gingival sulcus or the tooth pulp. mechanism.

Integrin and ADM: Integrin binding speciﬁcity regulates bio-

material surface chemistry effects on cell differentiation [57].

4.6. Mechanistic aspects In the present study, the groups that were most cured (speciﬁ-

cally the V-TW-5, P-TW-5) demonstrated over a 2-fold increase

Inﬂammation markers: The mRNA levels of tested inﬂamma- in gene expression supporting the favorable attachment of

tory markers IL-1, IL-6, IL-8, and TNF-␣ reﬂected the cytotoxic gingival ﬁbroblasts towards highly cured resin composites.

potential in both culture systems (Fig. 8 and 11) and were Adrenomedullin (ADM) has also previously been shown to

upregulated along with increasing cytotoxicity especially on increase ﬁbroblast-like synoviocyte adhesion to extracellular

low DC specimens. On the protein level, IL-6 concentrations matrix proteins by upregulating integrin activation [58]. In the

were elevated after direct exposure of cells to low DC RBC- present study, no differences in gene expression of ADM were

V specimens. Generally, inﬂammation marker upregulation observed between all groups. It may, therefore, be assumed

depended upon the RE and the DC. that ADM does not play a role in gingival ﬁbroblast attachment

Upregulation of IL-6 and IL-8 has also been observed, when to RBCs.

3-D-cultures of reconstituted human epithelial cells (cancer Apoptosis markers (caspase 9, and annexin A5) were sig-

cell line) had been exposed to non-toxic concentrations of niﬁcantly upregulated in direct cell material contact assays

TEGDMA, which is a commonly used acrylic monomer in RBCs as well as (not signiﬁcantly) caspases 3 and 8. This further

[43]. In a further study a UDMA based endodontic sealer caused conﬁrmed the relevance to more cytotoxic behavior, whereas

in pulp cells, PDL-cells and osteoblasts a signiﬁcant upreg- these gene expressions were not affected in an elution culture

ulation of inﬂammation markers IL-6, IL-8, IL-12 and TNF-␣ system. The resin monomer substances induced apoptosis in

parallel to a decrease in cell numbers. Also, upregulation of several cell types through the generation of redox balance-

IL-6 after exposure of human gingival ﬁbroblasts to extracts disturbing reactive oxygen species (ROS), which results in DNA

from chemically cured dental polymethacrylate materials has damage, followed by cell apoptosis [5,42,59].

been reported [44,45]. Interestingly, the combination of RBC-P and TW displayed

In clinical studies, gingival sulcus ﬂuid had been collected comparatively high mRNA expressions at the lowest radiant

␣

after the placement of subgingival Class V or II restorations exposure for IL-6 and TNF- and a signiﬁcant decrease of PDGF

using different materials [46,47]. With all materials concen- mRNA expression (growth factor for cell migration, prolifera-

trations of IL-6, IL-8 and TNF-␣ were higher than in control tion, and angiogenesis). Other genetic markers revealed some

teeth with no materials. However, this increase may be due to interesting tendencies such as the upregulation of Ki-67, cas-

several reasons, for example, plaque accumulation. Upregula- pase 9 and the downregulation of PDGF, but these were not

tion of inﬂammation mediators in the early phase of exposure signiﬁcantly different from controls.

to material in the gingival sulcus may interfere with the initial

healing process. Inﬂammation usually causes slower healing 4.7. Clinical relevance

[48]. Thus, increasing the inﬂammatory response during the

early healing phase might have an adverse effect on wound The radiant exposure and the DC of RBCs inﬂuence their bio-

healing [48]. logical behavior in a dose-dependent manner that is mainly a

Ki-67: In the present study mRNA expression of Ki-67 in logarithmic function. Consequently, in low RE situations, even

cells with direct contact to the test materials with low DC a small difference in the RE can cause a large difference in the

was decreased. This is in line with data from immunostaining cell reaction. Conversely, little effect is observed at higher REs.

with Ki-67 antigens of mouse ﬁbroblasts, were also reduced Biological effects are more pronounced in situations where

expression of Ki-67 after exposure to more than 10 mM Bis- there is direct cell contact to the RBC than when the effect

GMA was observed [49]. Schweikl et al. reported that the of the eluates is examined. Cell deaths, as well as induction

acrylic monomer TEGDMA caused cell cycle delays through of inﬂammation, apoptosis and reduced cell proliferation, are

p53-dependent and independent pathways in the various cell expected to occur after material placement. This local effect

lines [50]. This indicates a reduction of cell proliferation and may interfere with the tissue healing if the gingival tissues

can be seen in the context of the above-mentioned increase of have been traumatized during the tooth preparation or when

inﬂammation markers [48]. Thus, the initial healing potential ﬁlling the cavity.

d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193 1191

r e f e r e n c e s

Thus, it is reasonable to establish a critical minimum RE

that the RBC should receive based on the DC, biological reac-

tions or mechanical properties. Based on the DC, it appears

that the minimum RE should be between 15 to 25 J/cm for [1] Gaengler P, Hoyer I, Montag R. Clinical evaluation of

the two RBCs tested. For cellular reactions, the situation is posterior composite restorations: the 10-year report. J Adhes

Dent 2001;3:184–94.

more complicated, because these data depend not only on DC

[2] Brunthaler A, König F, Lucas T, Sperr W, Schedle A. Longevity

but also on a number of modifying parameters. Based on data

of direct resin composite restorations in posterior teeth: a

from cell attachment tests (Fig. 7) 100% cell attachment does

review. Clin Oral Investig 2003;7:63–70.

not occur after 3 elution cycles, even at a high DC, but does

[3] Federation FWD. FDI policy statement on dental amalgam

occur after 10 elution cycles, at which point this DC causes and the Minamata Convention on Mercury: adopted by the

almost no cell reaction. Thus, to prevent adverse local effects FDI General Assembly: 13 September 2014, New Delhi, India.

it is suggested that a radiant exposure of more than 25 J/cm Int Dent J 2014;64:295–6.

[4] Harlow J, Rueggeberg F, Labrie D, Sullivan B, Price R.

is advisable.

Transmission of violet and blue light through conventional

Photo-curing a 2 mm thick RBC that includes photoactiva-

(layered) and bulk cured resin-based composites. J Dent

tors that require violet light (RBC-P) using an LCU that emits

2016;53:44–50.

only blue light yielded inferior results, both for DC and in the

[5] Schmalz G, Arenholt-Bindslev D. Biocompatibility of dental

biological tests. Therefore, it is advisable that the emission materials. Springer; 2009.

proﬁle of the LCU should include the wavelengths of light [6] Cardoso KA, Zarpellon DC, Madruga CF, Rodrigues JA, Arrais

that cover the absorbance proﬁle of the initiators used in the CA. Effects of radiant exposure values using second and

RBC. third generation light curing units on the degree of

conversion of a lucirin-based resin composite. J Appl Oral

Sci 2017;25:140–6.

5. Conclusions [7] Geurtsen W. Substances released from dental resin

composites and glass ionomer cements. Eur J Oral Sci

1998;106:687–95.

Cell attachment/viability of human gingival ﬁbroblasts in con-

[8] Inoue K, Hayashi I. Residual monomer (Bis-GMA) of

tact with the RBCs or their eluates generally depended on the

composite resins. J Oral Rehabil 1982;9:493–7.

degree of conversion, the material composition and the num-

[9] Spahl W, Budzikiewicz H, Geurtsen W. Determination of

ber of eluate cycles (elution time). Cell reactions were more

leachable components from four commercial dental

pronounced in direct contact with test specimens than when composites by gas and liquid chromatography/mass

eluates were tested, and thus the local healing processes of spectrometry. J Dent 1998;26:137–45.

[10] Van Landuyt K, Nawrot T, Geebelen B, De Munck J,

the gingival tissues may be impaired by insufﬁcient curing.

Snauwaert J, Yoshihara K, et al. How much do resin-based

The RBC matrix components of RBC-V showed greater cell

dental materials release? A meta-analytical approach. Dent

attachment when compared to RBC-P, and eluates from RBC-

Mater 2011;27:723–47.

V demonstrated lower cell viability when compared to RBC-P.

[11] Ferracane J. Elution of leachable components from

Furthermore, the dual emission peak LCU that delivered vio- composites. J Oral Rehabil 1994;21:441–52.

let as well as blue light, especially for the acute effects (low [12] Sevkusic M, Schuster L, Rothmund L, Dettinger K, Maier M,

number of elutions), resulted in better cell attachment and Hickel R, et al. The elution and breakdown behavior of

constituents from various light-cured composites. Dent

less eluate cytotoxicity due to the presence of the additional

Mater 2014;30:619–31.

violet light. It was concluded that the inclusion of the violet

[13] Högg C, Maier M, Dettinger-Maier K, He X, Rothmund L, Kehe

light from the T2W was beneﬁcial, and the combination of

K, et al. Effect of various light curing times on the elution of

the dual emission peak T2W LCU, RBC-V, exposure delivering

composite components. Clin Oral Investig 2016;20:2113–21.

more than 25 J/cm reached the greatest DC that also had the [14] Pearson G, Longman C. Water sorption and solubility of

least cell response. resin-based materials following inadequate polymerization

by a visible-light curing system. J Oral Rehabil 1989;16:57–61.

[15] Sigusch BW, Pﬂaum T, Völpel A, Schinkel M, Jandt KD. The

Acknowledgements inﬂuence of various light curing units on the cytotoxicity of

dental adhesives. Dent Mater 2009;25:1446–52.

[16] Sigusch BW, Volpel A, Braun I, Uhl A, Jandt KD. Inﬂuence of

This work is supported in part by Kulzer GmbH, Hanau,

Germany. different light curing units on the cytotoxicity of various

dental composites. Dent Mater 2007;23:1342–8.

[17] Wegehaupt FJ, Lunghi N, Belibasakis GN, Attin T. Inﬂuence of

light-curing distance on degree of conversion and

Appendix A. Supplementary data

cytotoxicity of etch-and-rinse and self-etch adhesives. BMC

Oral Health 2016;17:12.

Supplementary material related to this article can be [18] Michaud PL, Price RB, Labrie D, Rueggeberg FA, Sullivan B.

found, in the online version, at doi:https://doi.org/10.1016/j. Localised irradiance distribution found in dental light curing

dental.2019.05.015. units. J Dent 2014;42:129–39.

1192 d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193

[19] Anselme K. Osteoblast adhesion on biomaterials. gordonii adhesion to dental material surfaces. Dent Mater

Biomaterials 2000;21:667–81. 2013;29:1080–9.

[20] Schweikl H, Müller R, Englert C, Hiller K-A, Kujat R, Nerlich [38] Germscheid W, de Gorre LG, Sullivan B, O’Neill C, Price RB,

M, et al. Proliferation of osteoblasts and ﬁbroblasts on model Labrie D. Post-curing in dental resin-based composites. Dent

surfaces of varying roughness and surface chemistry. J Mater 2018;34:1367–77.

Mater Sci Mater Med 2007;18:1895–905. [39] Müller R, Ruhl S, Hiller KA, Schmalz G, Schweikl H. Adhesion

[21] Schweikl H, Hiller K-A, Bolay C, Kreissl M, Kreismann W, of eukaryotic cells and Staphylococcus aureus to silicon

Nusser A, et al. Cytotoxic and mutagenic effects of dental model surfaces. J Biomed Mater Res A 2008;84:

composite materials. Biomaterials 2005;26:1713–9. 817–27.

[22] Setzer B, Bächle M, Metzger MC, Kohal RJ. The [40] Janus J, Fauxpoint G, Arntz Y, Pelletier H, Etienne O. Surface

gene-expression and phenotypic response of hFOB 1.19 roughness and morphology of three nanocomposites after

osteoblasts to surface-modiﬁed titanium and zirconia. two different polishing treatments by a multitechnique

Biomaterials 2009;30:979–90. approach. Dent Mater 2010;26:416–25.

[23] Zreiqat H, Howlett CR. Titanium substrata composition [41] Urcan E, Haertel U, Styllou M, Hickel R, Scherthan H, Reichl

inﬂuences osteoblastic phenotype: in vitro study. J Biomed FX. Real-time xCELLigence impedance analysis of the

Mater Res A 1999;47:360–6. cytotoxicity of dental composite components on human

[24] Schulz SD, Rüppell C, Tomakidi P, Steinberg T, Reichl F-X, gingival ﬁbroblasts. Dent Mater 2010;26:51–8.

Hellwig E, et al. Gene expression analysis of conventional [42] Goldberg M. In vitro and in vivo studies on the toxicity of

and interactive human gingival cell systems exposed to dental resin components: a review. Clin Oral Investig

dental composites. Dent Mater 2015;31:1321–34. 2008;12:1–8.

[25] Urcan E, Scherthan H, Styllou M, Haertel U, Hickel R, Reichl [43] Schmalz G, Schweikl H, Hiller KA. Release of prostaglandin

F-X. Induction of DNA double-strand breaks in primary E2, IL-6 and IL-8 from human oral epithelial culturemodels

gingival ﬁbroblasts by exposure to dental resin composites. after exposure to compounds of dental materials. Eur J Oral

Biomaterials 2010;31:2010–4. Sci 2000;108:442–8.

[26] Tadin A, Marovic D, Galic N, Kovacic I, Zeljezic D. [44] Diomede F, Caputi S, Merciaro I, Frisone S, D’arcangelo C,

Composite-induced toxicity in human gingival and pulp Piattelli A, et al. Pro-inﬂammatory cytokine release and cell

ﬁbroblast cells. Acta Odontol Scand 2014;72:304–11. growth inhibition in primary human oral cells after

[27] Styllou M, Reichl F-X, Styllou P, Urcan E, Rothmund L, Hickel exposure to endodontic sealer. Int Endod J 2014;47:

R, et al. Dental composite components induce DNA-damage 864–72.

and altered nuclear morphology in gingiva ﬁbroblasts. Dent [45] Borelli B, Zarone F, Rivieccio V, Riccitiello F, Simeone M,

Mater 2015;31:1335–44. Sorrentino R, et al. Polyacrylic resins regulate transcriptional

[28] Schweikl H, Hiller KA, Bolay C, Kreissl M, Kreismann W, control of interleukin-6, gp80, and gp130 genes in human

Nusser A, et al. Cytotoxic and mutagenic effects of dental gingival ﬁbroblasts. J Oral Sci 2017;59:87–91.

composite materials. Biomaterials 2005;26:1713–9. [46] Ilday NO, Celik N, Dilsiz A, Alp HH, Aydin T, Seven N, et al.

[29] Wang Y, Zhang Y, Jing D, Shuang Y, Miron RJ. Enamel matrix The effects of silorane composites on levels of cytokines and

derivative improves gingival ﬁbroblast cell behavior cultured periodontal parameters. Contemp Clin Dent 2013;4:

on titanium surfaces. Clin Oral Investig 2016;20:685–95. 437–42.

[30] Kobayashi E, Fujioka-Kobayashi M, Sculean A, Chappuis V, [47] Celik N, Askin S, Gul MA, Seven N. The effect of restorative

Buser D, Schaller B, et al. Effects of platelet rich plasma (PRP) materials on cytokines in gingival crevicular ﬂuid. Arch Oral

on human gingival ﬁbroblast, osteoblast and periodontal Biol 2017;84:139–44.

ligament cell behavior. BMC Oral Health 2017;17:91. [48] Martin P, Leibovich SJ. Inﬂammatory cells during wound

[31] Moharamzadeh K, Van Noort R, Brook IM, Scutt AM. repair: the good, the bad and the ugly. Trends Cell Biol

Cytotoxicity of resin monomers on human gingival 2005;15:599–607.

ﬁbroblasts and HaCaT keratinocytes. Dent Mater [49] Saraﬁan V, Uzunova Y, Hayrabedyan S, Ganchevska P,

2007;23:40–4. Filipova M, Filipov I, et al. Histo-blood group antigen

[32] Cuevas-Suarez CE, Pimentel-Garcia B, Rivera-Gonzaga A, expression and proliferative activity of ﬁbroblasts treated

Alvarez-Gayosso C, Ancona-Meza AL, Grazioli G, et al. with dental monomers. Cell Biol Toxicol 2008;24:27–37.

Examining the Effect of Radiant Exposure on Commercial [50] Schweikl H, Altmannberger I, Hanser N, Hiller K-A, Bolay C,

Photopolimerizable Dental Resin Composites. Dent J (Basel) Brockhoff G, et al. The effect of triethylene glycol

2018;6:55. dimethacrylate on the cell cycle of mammalian cells.

[33] Calheiros FC, Daronch M, Rueggeberg FA, Braga RR. Degree of Biomaterials 2005;26:4111–8.

conversion and mechanical properties of a BisGMA:TEGDMA [51] Schmalz G, Widbiller M, Galler KM. Signaling molecules and

composite as a function of the applied radiant exposure. J pulp regeneration. J Endod 2017;43:S7–11.

Biomed Mater Res B Appl Biomater 2008;84:503–9. [52] Ashcroft GS, Yang X, Glick AB, Weinstein M, Letterio JJ, Mizel

[34] Reichl F-X, Esters M, Simon S, Seiss M, Kehe K, Kleinsasser DE, et al. Mice lacking Smad3 show accelerated wound

N, et al. Cell death effects of resin-based dental material healing and an impaired local inﬂammatory response. Nat

compounds and mercurials in human gingival ﬁbroblasts. Cell Biol 1999;1:260–6.

Arch Toxicol 2006;80:370–7. [53] Jinno K, Takahashi T, Tsuchida K, Tanaka E, Moriyama K.

[35] Ferracane J, Marker V. Solvent degradation and reduced Acceleration of palatal wound healing in Smad3-deﬁcient

fracture toughness in aged composites. J Dent Res mice. J Dent Res 2009;88:757–61.

1992;71:13–9. [54] Pauly K, Fritz K, Furey A, Lobner D. Insulin-like growth factor

[36] Rothmund L, Shehata M, Van Landuyt KL, Schweikl H, Carell 1 and transforming growth factor-␤ stimulate

T, Geurtsen W, et al. Release and protein binding of cystine/glutamate exchange activity in dental pulp cells. J

components from resin based composites in native saliva Endod 2011;37:943–7.

and other extraction media. Dent Mater 2015;31: [55] Eckhardt A, Gerstmayr N, Hiller K-A, Bolay C, Waha C,

496–504. Spagnuolo G, et al. TEGDMA-induced oxidative DNA damage

[37] Schweikl H, Hiller K-A, Carl U, Schweiger R, Eidt A, Ruhl S, and activation of ATM and MAP kinases. Biomaterials

et al. Salivary protein adsorption and Streptococccus 2009;30:2006–14.

d e n t a l m a t e r i a l s 3 5 ( 2 0 1 9 ) 1173–1193 1193

[56] Schweikl H, Spagnuolo G, Schmalz G. Genetic and cellular synoviocyte adhesion to extracellular matrix proteins by

toxicology of dental resin monomers. J Dent Res upregulating integrin activation. Arthritis Res Ther 2010;12:

2006;85:870–7. R190.

[57] Keselowsky BG, Collard DM, García AJ. Integrin binding [59] Krifka S, Hiller K-A, Spagnuolo G, Jewett A, Schmalz G,

speciﬁcity regulates biomaterial surface chemistry effects Schweikl H. The inﬂuence of glutathione on redox

on cell differentiation. Proc Natl Acad Sci 2005;102: regulation by antioxidant proteins and apoptosis in

5953–7. macrophages exposed to 2-hydroxyethyl methacrylate

[58] M-DA Kioon, Asensio C, Ea H-K, Uzan B, Cohen-Solal M, (HEMA). Biomaterials 2012;33:5177–86.

Lioté F. Adrenomedullin increases ﬁbroblast-like